The amygdala and cingulate cortex are importantly involved in the etiology of emotional disturbances such as chronic anxiety, panic attacks and schitzophrenia. Yet, detailed knowledge of how these areas malfunction to bring about emotional disturbance has been elusive. This project approaches these issues, beginning with two fundamental premises. First, an understanding of the etiology of emotional disturbance will require a basic understanding of the role of the involved brain areas in production and control of emotional states. Second, individual brain areas do not influence psychological states in isolation. Instead, neural mediation of emotion and behavior emerges from the interactions of neurons in multiple brain areas that form functional circuits. This project continues an extensive programmatic analysis of the interactions of cingulate cortical and amygdalar neurons, and neurons in related areas, involved in mediation of fear-based learning of discriminative goal-directed behavior. The interactions are studied by recording the activity of neurons simultaneously in multiple brain sites during learning in rabbits. Selective circuit lesions and neurochemical manipulations are used to alter experimentally fear- and behavior-relevant information flow in the circuits. Recent discoveries set the stage for major advances concerning the documentation of specific learning-relevant brain pathways and circuit interactions. The new data demonstrate that amygdala neurons initiate rapidly-developing learning-related discriminative changes in the medial division of the medial geniculate nucleus (Mgm). The changes initiated in the Mgm enhance the sensory transmission of associatively significant stimuli, thereby facilitating behavioral performance. Intriguingly, the initiation process is blocked by briefly inactivating amygdala neurons during early training trials. If this is done, Mgm neurons can no longer develop these changes even after the inactivation is removed. The proposed research defines precisely the training period when the critical initiating events occur. Additional studies: test the hypothesis that amygdalar efferents act via synapses in auditory cortex to initiate the critical changes in the Mgm and establish the particular amygdalar subdivisions involved in the initiation process. The recent data also show that amygdala efferents are essential for the establishment of learning-related changes in the cingulate cortex and limbic thalamus. The changes in these areas are essential for acquisition of the learned behavioral response. Two functionally distinct cingulothalamic circuits exhibit early neuronal changes that mediate the initial stages of behavioral learning and late neuronal changes that mediate the performance of the learned behavior at asymptotic levels. Amygdalar efferents are essential for the early and the late cingulothalamic changes. These two circuits receive input from different amygdalar nuclei. The proposed studies test the hypotheses that neurons in the different nuclei exhibit the early and late forms of learning-related activity respectively and are the specific amygdalar nuclei involved respectively in production of the early and late cingulothalamic activity.